Original title: Bayes Estimation of Queue Length
Translated title: Bayesovské odhadování délky kolony
Authors: Dohnal, Pavel
Document type: Papers
Conference/Event: International PhD Workshop on Interplay of Societal and Technical Decision-Making, Young Generation Viewpoint /7./, Hrubá Skála (CZ), 2006-09-25 / 2006-09-30
Year: 2006
Language: eng
Abstract: [eng] [cze]
The paper deals with an application of Bayes for estimation of the queue length in junction arm. In Bayesian view the concept of probability is not interpreted in terms of limits of relative frequencies but more generally as a subjective measure of belief of a rationally and consistently reasoning person which is used to describe quantitatively the uncertain relationship between the statistician and the external world. This model splits controlled networks into microregions. The queue length and the occupancy of each junction approach are the basic state quantities for fully expressed traffic situation at given time instant. The occupancy determines relative time of the detector activation during sample period, i.e. the proportion of time when detector has been occupied and total time of measuring period. The optimization criterion for this attitude is minimization of the queue length. For clearness, the model is derived for simple junction. Příspěvek se zabývá aplikací Bayesovského přístupu na odhadování délek kolon na jednotlivých ramenech křižovatek.
Keywords: Bayes estimation; intensity; occupancy; queue length; traffic flow
Project no.: CEZ:AV0Z10750506 (CEP), 1M0572 (CEP), 1ET100750401 (CEP)
Funding provider: GA MŠk, GA AV ČR
Host item entry: Proceedings of the 7th International PhD Workshop: Interplay of Societal and Technical Decision-Making, Young Generation Viewpoint

Institution: Institute of Information Theory and Automation AS ČR (web)
Document availability information: Fulltext is available at the institute of the Academy of Sciences.
Original record: http://hdl.handle.net/11104/0136807

Permalink: http://www.nusl.cz/ntk/nusl-35789


The record appears in these collections:
Research > Institutes ASCR > Institute of Information Theory and Automation
Conference materials > Papers